Acoustic Emission Monitoring in Additive Manufacturing

Acoustic emission (AE) monitoring has emerged as a powerful non-destructive technique for real-time assessment of additive manufacturing (AM) processes. As high-energy sources such as lasers or electron beams interact with powder or wire feedstocks, elastic waves are generated by rapid thermal expansion, phase changes and dynamic melt-pool phenomena. These emissions, captured by airborne or structure-borne sensors, carry rich information on defect initiation, pore formation and transitions between processing regimes. By analysing waveform features, spectrograms and signal statistics, researchers can distinguish conduction, keyhole and lack-of-fusion modes and detect anomalies such as spatter, microcracks and pore onset. Integration of AE monitoring into AM platforms supports closed-loop control, enabling process adjustments and build termination to safeguard part quality. Beyond defect detection, AE analysis contributes to the development of processing maps, transfer-learning strategies for different materials and domain-adaptation schemes to enhance model robustness. With its capacity for high temporal resolution, minimal hardware footprint and potential for cost-effective implementation, AE monitoring addresses critical industrial demands for repeatable, reproducible and scalable AM production.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Acoustic Emission Monitoring in Additive Manufacturing publication trend

The graph below shows the total number of articles in acoustic emission monitoring in additive manufacturing across all publications each year (not limited to Nature Index journals).

Technical terms

Acoustic emission (AE): Elastic waves generated by rapid mechanical or thermal events during material processing, captured for non-destructive evaluation.

Laser powder bed fusion (LPBF): An AM technique wherein a laser selectively fuses successive layers of metal powder to build parts.

Keyhole mode: A deep, vapour-stabilised melt-pool regime characterised by high energy density and potential pore formation due to plasma-driven recoil pressure.

Lack-of-fusion: A defect regime arising from insufficient energy input, leading to incomplete particle melting and interlayer bonding.

Conduction mode: A melt-pool regime in which heat conduction predominates and the pool remains shallow and stable.

Processing map: A graphical representation of regimes and defect thresholds across combinations of power and scan speed.

Domain adaptation: A machine-learning technique that aligns feature distributions between differing data sources to enhance model generalisability.

References

1. Self-Supervised Bayesian representation learning of acoustic emissions from laser powder bed Fusion process for in-situ monitoring. Materials & Design (2023).
2. Acoustic emission for the prediction of processing regimes in Laser Powder Bed Fusion, and the generation of processing maps. Additive Manufacturing (2023).
3. Monitoring of Laser Powder Bed Fusion process by bridging dissimilar process maps using deep learning-based domain adaptation on acoustic emissions. Additive Manufacturing (2024).